NEUROIMAGING OF ARACHIDONIC ACID PARTICIPATION IN BRAIN SIGNAL TRANSDUCTION.[unreadable] We used our fatty acid method to image brain arachidonic acid (AA, 20:4n-6) signaling in unanesthetized rats administered a cholinergic muscarinic agonist, arecoline. In rats pretreated with the non-selective cyclooxygenase (COX) inhibitor flurbiprofen, and in mice in which cyclooxygenase (COX)-2 was knocked out, we showed that the neuroreceptor-initiated AA signal, which is mediated by activation of cytosolic phospholipase A2 (cPLA2) to release from AA from membrane phospholipid, represents metabolic loss of AA specifically via cyclooxygenase (COX)-2 (1). During functional activation and at baseline, AA incorporation into brain from plasma equals the rate of its metabolic loss within brain. The COX-2 knockout condition in the mouse produces multiple changes in brain lipid composition (2).[unreadable] [unreadable] DOPAMINE RECEPTOR SIGNALING INVOLVING ARACHIDONIC ACID. [unreadable] Post-synaptic dopaminergic D2 but not D1 receptors are coupled to cytosolic cPLA2 activation and AA release from membrane phospholipid. We showed that D-amphetamine, a drug used to treat hyperactive children and a cause of addiction in adults, stimulated AA signaling in rat brain by reversing the direction of the pre-synaptic dopamine reuptake transporter (DAT) and increasing synaptic dopamine so as to stimulate dopamine D2 receptors. This effect of amphetamine could be blocked by pretreatment with the D2 receptor antagonist, raclopride. The ability to image effects of drugs on D2-receptor mediated signal transduction involving AA may allow us to use PET to determine where in the brain and to what extent this signaling is altered with aging and in patients Alzheimer disease, Parkinson disease, or attention deficit hyperactivity disorder (ADHD) (3).[unreadable] [unreadable] DISTURBED DOPAMINE SIGNALING IN RAT MODEL OF PARKINSON DISEASE. [unreadable] Parkinson disease involves loss of dopamine-producing neurons in the substantia nigra, and fewer pre-synaptic dopamine transporters (DATs) and more post-synaptic D2 receptors in terminal areas of these neurons. We prepared rats with a chronic unilateral lesion of the substantia nigra as a model of Parkinson disease. In these rats, the AA signal in response to D-amphetamine, which reverses the direction of the DAT, was reduced in terminal areas ipsilateral to the lesion, whereas the signal was increased in rats given the D2 receptor agonist, quinpirole. Baseline (non-drug) AA incorporation was increased in ipsilateral areas, as was the activity of cPLA2. The increased baseline AA incorporation and increased AA responsiveness to quinpirole is consistent with higher ipsilateral D2 receptor densities and cPLA2 activity levels, whereas reduced responsiveness to D-amphetamine is consistent with dropout of pre-synaptic elements containing the DAT. Noninvasive imaging of AA signaling with appropriate dopaminergic drugs thus might be used with PET to identify pre- and post-synaptic changes in the course of Parkinson and other diseases in human subjects (4).[unreadable] [unreadable] DOWNREGULATION OF GLUTAMATERGIC NEUROTRANSMISSION VIA ARACHIDONIC ACID BY CARBAMAZEPINE. [unreadable] There is genetic and postmortem evidence for disturbed glutamatergic neurotransmission in bipolar disorder, but studies have not examined whether drugs effective against the bipolar disorder might rectify the disturbance. We showed that chronic administration to rats of carbamazepine, an anticonvulsant effective against mania in bipolar disorder, reduced arachidonic acid signal in brain in response to a subconvulsant dose of the glutamatergic N-methyl-D-aspartate (NMDA) receptor agonist, NMDA (5). This result agrees with our prior publications that chronic lithium and valproic acid did so as well. As NMDA is a glutamate receptor allowing calcium into the cell to activate cPLA2, these data taken together indicate that anti-manic drugs generally downregulate glutamatergic neurotransmission. The data are consistent with evidence for disturbed glutamatergic and NMDA signaling in the bipolar disease brain.[unreadable] [unreadable] CHRONIC LITHIUM ATTENUATES UPREGULATED BRAIN ARACHIDONIC ACID METABOLISM IN A RAT MODEL OF NEUROINFLAMMATION.[unreadable] We had reported that neuroinflammation, caused by a 6-day intra-cerebroventricular infusion of bacterial lipopolysaccharide in rats, is associated with upregulation of markers of brain arachidonic acid (AA) metabolism, including increased incorporation of AA from plasma on neuroimaging and increased brain cPLA2 activity and eicosanoid concentrations. We now found that chronic pretreatment with a therapeutically relevant dose of lithium, a drug used to treat bipolar disorder, prevented or reduced these inflammatory changes. As an upregulated arachidonic cascade likely has neuropathological consequences, lithium might be considered for treating human brain diseases accompanied by neuroinflammation, such as Alzheimer disease (6).[unreadable] [unreadable] CHRONIC LITHIUM ADMINISTRATION TO RATS ELEVATES GLUCOSE METABOLISM IN WIDE AREAS OF BRAIN. [unreadable] The regional cerebral metabolic rate for glucose (rCMRglc) can be imaged in awake animals using quantitative autoradiography, after injecting radiolabeled 2-deoxy-D-glucose intravenously. rCMRglc is a marker of brain energy metabolism and functional activity. We reported that rCMRglc was widely elevated in the brain of unanesthetized rats treated chronically with lithium at a dose relevant to bipolar disorder (3). This elevation, not examined previously in animals or humans, may relate to lithium's reported ability to increase auditory and visual evoked responses in healthy subjects and to cause convulsions at toxic doses in bipolar patients, as well as to lower the convulsive threshold to cholinergic drugs in rats (7).[unreadable] [unreadable] HIGH ATP CONSUMPTION BY BRAIN LIPID METABOLISM[unreadable] Until recently, lipid metabolism was considered to consume less than 2% of the brains ATP, and thus not to participate in active brain processes. In contrast, we calculated that de novo phospholipid synthesis, fatty acid recycling in phospholipids, phosphatidylinositide turnover, and maintenance of membrane asymmetries of ether phospholipids, together consume about 25% of the net ATP consumed by brain (8). This high rate of energy use is consistent with the very active contribution of lipid metabolism to brain function and structure.